Tenyks Blog

We bring you the main key takeaways of NVIDIA's GTC 2024.

How to increase model performance on any Roboflow model using Tenyks' Data-Centric approach.

In this third part, we show how Tenyks can help you in identifying model failures caused by data issues.

In this second part, we show you how to use the Tenyks API to upload your NVIDIA TAO trained model.

We short list, ordered by date, the go-to conferences for professionals looking to stay ahead, offering a platform to discuss cutting-edge research and real-world applications of AI.

In this opening post, we demystify the NVIDIA ecosystem and lay the ground to build a Data-Centric pipeline.

How Foundation Models are transforming the Computer Vision pipeline.

A guide to comparing multiple models in an NVIDIA TAO Toolkit object detection pipeline for edge computing, including common problems and solutions.

Learn about the last trends in synthetic data generation.

Switching from Model-Centric to Data-Centric Benchkmarks.

A quick setup guide for an NVIDIA TAO Toolkit (v3 & v4) object detection pipeline for edge computing, including tips & tricks and common pitfalls.

A brief overview of AI Safety, along with highlights of AI safety works presented at NeurIPS 2023.

We show you, step by step, how to set up your own Image-to-Image search tool.

We analyze an object detection dataset to find where it is dragging model performance.

We debunk the misconception that machine learning glory resides in one model.

We debug, step by step, a manufacturing casting defects dataset for object detection.

We breakdown ControlNet, one of ICCV23 best paper winners, to understand how you can control Diffusion Models' output.

ICCV23 workshop "Quo Vadis, Computer Vision?" answers this query.

Your current data selection process may be limiting your models.

A practical blueprint of what skills you need to master to become a proficient computer vision engineer in 2023.

Unlock Insights from Unlabelled Images in Machine Learning: Exploring Object Detection Datasets with and without Labels.

Use embeddings with Vector DBs to perform multi-modal search on images.

Learn about the Foundation Models - for object classification, object detection, and segmentation - that are redefining Computer Vision.

We break down the 5 most crucial & hidden challenges to achieve autonomous driving. Disclaimer: It is a data problem as much as a technological one, from what we learned at CVPR workshops on SOTA AV.

We unveil what Vector DBs are, why they are useful, and how to use them.

We breakdown the responsibilities for ML Engineers & MLOps Engineers highlighting where they overlap and where they differ.

Tackle occlusion, label quality, data imbalance and scale variation in industry-level agriculture datasets for object detection.

Be aware of this 4 main issues in agriculture datasets for object detection.

Improve your object detection model performance with these five tips on how to deal with data-quality and model failures.

The Multiclass Confusion Matrix for Object Detection enables ML Engineers to identify edge cases, imbalance issues and other failures in your dataset.

Machine Learning Operations (MLOps) have become increasingly important in recent years as organizations look to integrate machine learning models into their operations. In this blog post, we will discuss the Top 15 MLOps and Computer Vision conferences for computer vision engineers to attend in 2023!

MLOps, CVOps, & Artificial Intelligence Events in 2023

We breakdown and demystify common object detection metrics, including mean average precision (mAP) and mean average recall (mAR).